Researchers have developed a new protocol for generating and evaluating codebooks using large language models (LLMs) to assess the quality of parliamentary discourse. This method aims to overcome the limitations of traditional expert-scored codebooks, which are rigid and prone to bias. The proposed protocol compares top-down, bottom-up, and hybrid approaches for schema generation, introducing intrinsic inter-model agreement (IMA) and extrinsic proxy measures for evaluation. Experiments on Israeli parliamentary transcripts showed that data-driven schemas achieved higher IMA and maintained validity, suggesting their effectiveness for large-scale political discourse analysis. AI
IMPACT This research offers a more efficient and less biased method for analyzing political discourse, potentially improving the study of legislative engagement.
RANK_REASON The cluster contains a research paper detailing a new methodology for using LLMs in discourse analysis. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
